Graham McNicoll

I cannot look at 100 experiments and tell you what they have in common. That is the problem an experimentation program hits as it scales. The number of tests easily moves past what one person can hold in their head, so the conclusions and learnings sit there unread/unused because no one can read them all and identify patterns. An agent can. We have built a full set of skills that teach your agents how to use GrowthBook and pull in past experiments, so when you ask one to build a feature, it checks what has already been tried first. The result is that you'll build better products faster by leveraging what you've learned. The write-up is at https://lnkd.in/gwm_9ScV